2020-02359 - Post-Doctoral Research Visit F/M Learning controllable representations that evolve over time

Contract type : Fixed-term contract

Level of qualifications required : PhD or equivalent

Fonction : Post-Doctoral Research Visit

About the research centre or Inria department

The Inria Lille - Nord Europe Research Centre was founded in 2008 and employs a staff of 360, including 300 scientists working in sixteen research teams. Recognised for its outstanding contribution to the socio-economic development of the Hauts-De-France région, the Inria Lille - Nord Europe Research Centre undertakes research in the field of computer science in collaboration with a range of academic, institutional and industrial partners.

 The strategy of the Centre is to develop an internationally renowned centre of excellence with a significant impact on the City of Lille and its surrounding area. It works to achieve this by pursuing a range of ambitious research projects in such fields of computer science as the intelligence of data and adaptive software systems. Building on the synergies between research and industry, Inria is a major contributor to skills and technology transfer in the field of computer science.

Context

The Inria team SequeL is a very active, united, hard-working, internationally renowned and connected research team specialized on theoretical and applied aspects of machine learning for sequential decision making with noisy or partial feedback. It is focused on reinforcement, bandit learning, especially in non-stationary environments.

Our work spans from learning theory, to the design of efficient algorithms, to applications. Our team led to many publications in top conferences such as NeurIPS, ICML, ALT, COLT, AISTATS.

Assignment

In order to act, an agent should learn a representation of the world. Hopefully, the representation of states should be controllable: by learning a representation of the actions, the agent can act on the representations of states. In order for these representations to be meaningful, the actions should modify few independently controllable features of the representations.

For example, in educational assessments, we can learn the latent ability of students as we ask them questions. These latent abilities evolve over time, and lessons act on these representations.

As main outcome, we expect publications at top conferences and journals in machine learning or data mining.

Supervision: Jill-Jênn Vie

Skills

  • deep generative models
  • reinforcement learning
  • recommender systems

Experience in various areas is a plus :

  • privacy-preserving ML
  • fairness
  • causal inference
  • applications to education or healthcare.

Benefits package

  • Partial reimbursement of public transport costs
  • Subsidized meals
  • Leave : 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours)
  • Possibility of teleworking (after 6 months of employment) and flexible organization of working hours
  • Professional equipment available (videoconferencing, loan of computer equipment, etc.)
  • Access to vocational training
  • Possibility of French courses
  • Social, cultural and sports events and activities
  • Administrative support : Social security coverage/ Help for Housing / Scientific Resident card and help for visa

Remuneration

Gross monthly salary (before taxes) : 2653 €